[HACKERS] Typo in bufmgr.c that result in waste of memory

ivateRefCountEntry); PrivateRefCountHash = hash_create("PrivateRefCount", 100, &hash_ctl, HASH_ELEM | HASH_BLOBS); } Attached patch is for the latest development version at the git repository. -- Takashi Horikawa NEC Corporation Knowledge

Re: [HACKERS] Typo in bufmgr.c that result in waste of memory

I think we should fix it, but not backpatch. I agree that backpatch is unnecessary. I hope it will be fixed for the current development version at some suitable opportunity. Best regards, Takashi Horikawa NEC Corporation Knowledge Discovery Research Laboratories > -Original Message- >

Re: [HACKERS] patch: prevent user from setting wal_buffers over 2GB bytes

I have confirmed that applying the patch makes 'wal_buffers = 4GB' works fine, while original PostgreSQL-9.4.4 results in core dump (segfault). I'll be happy if anyone reconfirm this. -- Takashi Horikawa NEC Corporation Knowledge Discovery Research Laboratories > -Origina

Re: [HACKERS] patch: prevent user from setting wal_buffers over 2GB bytes

Size) XLOG_BLCKSZ;' in GetXLogBuffer(XLogRecPtr ptr) and 'NewPage = (XLogPageHeader) (XLogCtl->pages + nextidx * (Size) XLOG_BLCKSZ); ' in AdvanceXLInsertBuffer(XLogRecPtr upto, bool opportunistic) , etc. Only exception is 'page = &XLogCtl->pages[firstIdx * XLOG_BLCKSZ];

[HACKERS] Partitioned checkpointing

SSDs for $PGDATA, for which seek optimizations are not believed effective. Therefore, this mechanism is worth to further investigation aiming to implement in future PostgreSQL. -- Takashi Horikawa NEC Corporation Knowledge Discovery Research Laboratories partitioned-checkpointing.patch Description: Binary data smime.p7s Description: S/MIME cryptographic signature

Re: [HACKERS] Partitioned checkpointing

_xlog > raid1 of 2 SSDs for $PGDATA (other than pg_xlog) 'raid1' is wrong and 'raid0' is correct. -- Takashi Horikawa NEC Corporation Knowledge Discovery Research Laboratories > -Original Message- > From: Alvaro Herrera [mailto:pgsql-hackers-ow...@postgres

Re: [HACKERS] Partitioned checkpointing

Fabien, Thanks for your comment. I'll check them and try to examine what is the same and what is different. -- Takashi Horikawa NEC Corporation Knowledge Discovery Research Laboratories > -Original Message- > From: pgsql-hackers-ow...@postgresql.org > [mailto:pg

Re: [HACKERS] Partitioned checkpointing

om bufmgr.h to xlog.h. It would be also trivial. Best regards. -- Takashi Horikawa NEC Corporation Knowledge Discovery Research Laboratories partitioned-checkpointing-v2.patch Description: Binary data smime.p7s Description: S/MIME cryptographic signature

Re: [HACKERS] Partitioned checkpointing

artition id = buffer id % number of partitions' is fairly simple. Best regards. -- Takashi Horikawa NEC Corporation Knowledge Discovery Research Laboratories > -Original Message- > From: Simon Riggs [mailto:si...@2ndquadrant.com] > Sent: Friday, September 11, 2015 10:57 PM

Re: [HACKERS] Partitioned checkpointing

I think it's likely that the patch will have only a very small effect if > applied ontop of Fabien's patch (which'll require some massaging I'm sure). It may be true or not. Who knows? I think only detail experimentations tell the truth. Best regards. -- Takashi Horikawa

Re: [HACKERS] Partitioned checkpointing

ing' case, the results shown in that graph is probably worth than bug-fix version. Best regards. -- Takashi Horikawa NEC Corporation Knowledge Discovery Research Laboratories > -Original Message- > From: pgsql-hackers-ow...@postgresql.org > [mailto:pgsql-hackers-ow...@postgres

Re: [HACKERS] Partitioned checkpointing

> As to 'partitioned checkpointing' case, the results shown in that graph > is probably worth than bug-fix version. ^ worse Sorry for typo. -- Takashi Horikawa NEC Corporation Knowledge Discovery Research Laboratories > -Original

Re: [HACKERS] Partitioned checkpointing

rtitioned checkpointing' is effective to eliminate (or mitigate) them; DBT-2 is an example. And also, 'Partitioned checkpointing' is worth to study why it is effective in pgbench executions. By studying it, it may lead to devising better ways. -- Takashi Horikawa NEC Corporation Kno